Only limited information is available concerning the glycosyl- plasmanylinositol (GPI) anchor preassembly pathway in mammalian cells, its relationship to biochemical reactions responsible for synthesis of non- anchor GPIs, and defect(s) in it which underlie deficient surface protein expression in experimental mutant cells and in naturally occurring paroxysmal nocturnal hemoglobinuria (PNH) affected blood elements. In preliminary studies we have 1) partially characterized glucosaminyl (GlcN) Pl precursors (designated GPI-A and -B) and mannosyl (man)n and substituted (X)-Man3-GlcN-Pls (designated GPI-C-G) with properties of GPI-anchor pathway intermediates, 2) partially analyzed abnormal products which are synthesized in Thy-1 lymphomas, 3) derived GPI-anchor defective human K562 cell mutants which exhibit defects different from those in the lymphomas, and 4) obtained information on the site of the biochemical lesion(s) responsible for PNH. Based on the data available so far, the mammalian GPIs resemble corresponding Trypanosoma brucei (Tryp) GPIs with respect to their glycan core structures but differ in that they are based on alkylglycerol and uniformly contain acylated (acyl) inositol (I). While only 4 of the previously described Thy-1 mutants synthesize GPIs and of these 2 (classes E and F) generate abnormal products, at least 1 K562 mutant (IVEE) synthesizes normal appearing precursors. Of 8 PNH patients examined to date, affected leukocytes of 7 show an ability to synthesize GlcN-PI (GPI-B) but not X-Man3-GlcN-PI (GPI-G). The proposed experiments are directed at 1) complete chemical characterization of normal GPI-anchor pathway intermediates, 2) localization of the subcellular sites of enzymatic activities associated with synthesis of GPI-anchor precursors, 3) derivation of additional K562 cell mutants (for GPI structural studies, localization of PNH defects, and rescue attempts via gene reconstitution), 4) development of assays for individual GPI-anchor synthetic reactions (for quantitating and purifying GPI-anchor enzymes), and 5) identification of genes encoding GPI-anchor enzymes via transfections employing antisense and sense RNA inhibition and reconstitution strategies. The data obtained should provide insights into GPI intracellular biosynthesis which will pertain to mammalian GPI-anchored proteins in general and have clinical as well as basic relevance.